CN111337042B - Vehicle path planning method and system - Google Patents
Vehicle path planning method and system Download PDFInfo
- Publication number
- CN111337042B CN111337042B CN202010176733.6A CN202010176733A CN111337042B CN 111337042 B CN111337042 B CN 111337042B CN 202010176733 A CN202010176733 A CN 202010176733A CN 111337042 B CN111337042 B CN 111337042B
- Authority
- CN
- China
- Prior art keywords
- antibody
- antibodies
- affinity
- forgetting
- vehicle path
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/26—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network
- G01C21/34—Route searching; Route guidance
- G01C21/3446—Details of route searching algorithms, e.g. Dijkstra, A*, arc-flags, using precalculated routes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biophysics (AREA)
- Theoretical Computer Science (AREA)
- Remote Sensing (AREA)
- Radar, Positioning & Navigation (AREA)
- General Physics & Mathematics (AREA)
- Computing Systems (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Molecular Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Genetics & Genomics (AREA)
- Physiology (AREA)
- Automation & Control Theory (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention belongs to the technical field of vehicle path planning and discloses a vehicle path planning method and a vehicle path planning system.A plurality of random paths are initialized to be used as antibodies to be put into an antibody candidate set, the distance of each vehicle path is calculated to be used as the affinity between each antibody and an antigen, and each antibody is initialized and marked; selecting the first n antibodies with highest affinity, cloning each antibody, and updating the marker; re-labeling the mutated antibody; recalculating affinity of the mutated antibodies, selecting N antibodies with highest affinity with the antigen, and updating antibody markers of the antibodies in the antibody candidate set; when the antibody meets an antibody forgetting threshold, forgetting; and terminating the calculation when the number of times of calling the evaluation function is met to obtain the optimal vehicle path. According to the invention, through a replacement process, the antibodies in the antibody set are subjected to population labeling and updating, so that the effects of improving the convergence speed and the convergence stability of the algorithm are achieved.
Description
Technical Field
The invention belongs to the technical field of vehicle path planning, and particularly relates to a vehicle path planning method and system.
Background
At present, the clonal selection algorithm is a novel intelligent optimization algorithm which is inspired by the clonal selection principle of a biological immune system; ag (antigen) antigen: the invention refers in particular to the requirements of vehicle path planning; ab (antibody) antibody: the invention refers in particular to each city path sequencing sequence; forgetting mechanism (forking mechanism): forgetting is a phenomenon of information loss, and the loss of information is significant under certain circumstances.
The Vehicle Routing planning Problem (VRP) was first proposed in 1959 by Dantzig and Ramser, which means that a certain number of customers, each with a different number of cargo demands, provide the customers with cargo by a distribution center, distribute the cargo by a fleet of vehicles, organize appropriate driving routes, and achieve the objectives of shortest route, minimum cost, minimum time consumption, etc. under certain constraints. Since the feasible solution to the problem is a full permutation of all vertices, as the number of vertices increases, combinatorial explosion occurs, which is an NP-complete problem.
As it has been widely used in the fields of transportation, circuit board line design, logistics distribution, etc., a great deal of research has been conducted by domestic and foreign scholars. The conventional solving method for the vehicle route problem can be divided into an exact algorithm (exact algorithm) and a heuristic solution (heuritics), wherein the exact algorithm comprises a branch boundary method, a branch cutting method, a set covering method and the like; the heuristic solution includes a saving method, a simulated annealing method, a deterministic annealing method, a tabu search method, a genetic algorithm, a neural network, an ant colonizing algorithm and the like. However, as the scale of the problem increases, the exact algorithm becomes ineffective, and therefore, in later studies, the prior art has focused on using an approximate algorithm or a heuristic algorithm. The method for solving the shortest path problem in the current main vehicle path planning problem and the defects thereof are as follows:
1. and (3) simulating an annealing algorithm: the convergence speed is low, the execution time is long, and the parameter dependence is large;
2. genetic algorithm: the method is easy to fall into local precocity, has poor convergence performance, and solves the combined problem from the continuous problem, so that the precision is greatly influenced;
3. clone selection algorithm: the method belongs to an evolutionary algorithm cluster, has high convergence rate and is easy to fall into local optimum; the clonal selection algorithm has two problems in solving the vehicle path planning problem: firstly, the intermediate antibody cannot be eliminated in time to influence the efficiency of the algorithm, and secondly, the newly generated antibody has insufficient competitiveness when the diversity of the antibody is ensured.
4. Ant colony algorithm: the calculation cost is too large, and the solving efficiency is not high.
Through the above analysis, the problems and defects of the prior art are as follows:
the existing vehicle path planning method has inaccurate result, low precision, low efficiency and long time consumption. If simulated annealing is used as a core algorithm to solve the vehicle planned path in the logistics vehicle path planning system, the method is low in convergence speed, so that a user needs to wait for a long time to obtain a solution, and the method is not friendly to the user and low in system efficiency. And when a genetic algorithm or an original clone selection algorithm is used as a core algorithm of the logistics vehicle path planning system, the method is easy to fall into local precocity, and the obtained result is often not an optimal path, so that manpower and material resources are consumed in actual logistics distribution, and the logistics distribution efficiency is reduced.
The difficulty in solving the above problems and defects is: and the accuracy of the operation result of the method is improved under the condition of ensuring the high convergence rate of the method. The user does not need to wait too long, and a good logistics vehicle path planning result can be obtained. Namely, the quality of the logistics vehicle path planning is improved on the premise of ensuring the time efficiency.
The significance of solving the problems and the defects is as follows: the convergence rate and stability of the method are improved, and the solving capability of the method is enhanced. The system performance for solving the logistics vehicle distribution path is improved, a user can obtain a more effective distribution path more quickly, vehicles can move the least distance under the condition that the demands are the same, the logistics distribution efficiency is improved, and manpower and material resources are saved.
Disclosure of Invention
Aiming at the problems in the prior art, the invention provides a vehicle path planning method and a vehicle path planning system.
The method is realized by applying the vehicle path planning method to a logistics vehicle path planning system as a core algorithm, and provides a vehicle path planning function for the system. The core algorithm comprises the following steps: initializing N random paths as antibodies, putting the random paths into an antibody candidate set, calculating the distance of each vehicle path as the affinity between each antibody and an antigen, and initializing and marking each antibody; selecting the first n antibodies with highest affinity, cloning each antibody, and updating the marker; re-labeling the mutated antibody; recalculating affinity of the mutated antibodies, selecting N antibodies with highest affinity with the antigen, and updating antibody markers of the antibodies in the antibody candidate set; when the antibody meets an antibody forgetting threshold, forgetting; and terminating the calculation when the number of times of calling the evaluation function is met to obtain the optimal vehicle path.
Further, the vehicle path planning method comprises the following steps:
acquiring relevant city number and coordinate data, initializing N antibodies, putting the N antibodies into an antibody candidate set, calculating the affinity between each antibody and an antigen, and initializing a marker;
step two, ordering the calculated affinities of the antigen and the antibody, selecting the first n antibodies with the highest affinities, cloning each antibody, and updating and marking the cloned antibodies;
step three, carrying out mutation operation on the cloned antibody, and updating and marking the cloned antibody;
step four, recalculating the affinity of the mutated antibodies, selecting N antibodies with the highest affinity with the antigen, putting the N antibodies into an antibody candidate set, and adding 1 to the survival time values of the N antibodies;
judging whether the antibody meets an antibody forgetting threshold, and if so, executing forgetting operation; if not, turning to the sixth step;
judging whether the algorithm meets the evaluation function calling times, and if so, outputting an optimal path, an optimal path graph and an optimal path distance; if not, returning to the step two.
Further, in the first step, the affinity between the antibody and the antigen is calculated as follows:
wherein citysize is the total number of cities the vehicle will pass, xj、yjIs the longitude and latitude, ab, of city jiIs an antibody. The specific process of initializing the labeling in step one is to assign an initial survival time value of 1 and an antibody competition attribute value of 1 to each antibody.
Further, in step two, the cloning comprises:
for antibody abiThe number of clones satisfies the following formula:
wherein a and b are both constants, and a >0, max _ clone is the maximum number of clones. And the specific process of updating the marker in the second step is to add 1 to the competitive attribute value of the cloned antibody.
Further, in step three, the variation includes:
the higher the affinity, the smaller the probability of antibody mutation, and mutation is carried out by using a mutation operator;
the mutation operator is:
among them, operator (sigma, ab)i) Is to antibody abiA certain city and neighbor cities with the interval of sigma are subjected to cross variation,
and sigma is taken according to the affinity of the antibody. The third stepThe new labeling process assigns an initial survival time value of 1 and an antibody competition attribute value of 1 to the antibody producing the variation.
Further, in step five, the forgetting includes:
the forgetting judgment is carried out, the forgetting degree of the antibody is calculated according to the current labeling state of the antibody, the antibody is compared with a forgetting threshold value, and when the antibody meets the forgetting judgment condition, the antibody is replaced by a new antibody; acquiring iteration times gen of a current algorithm, and carrying out gen-round clonal variation operation on a new antibody; and performing antibody initialization marking on the mature antibody selected after the operation, judging whether the number of times of calling the evaluation function is met, outputting the shortest path and the total length of the path when the number of times of calling the evaluation function is met, and returning to the step two if the number of times of calling the evaluation function is not met.
Further, in the fifth step, the forgetting formula is as follows:
wherein popsize is the antibody candidate population size, abnewIs a new antibody randomly generated from solution space, pre (-) is the antibody pre-treatment process, and c is the forgetting threshold. Calculating the degree of amnesia of the antibody
Among them, iteration (ab)i) Represents an antibody abiThe number of rounds participating in the iteration in the candidate antibody set, i.e. the survival time value, strength (ab)i) Represents an antibody abiA value of a competitive attribute in the current iteration antibody candidate set.
Another object of the present invention is to provide a system for planning a route using a vehicle, comprising:
the antibody and antigen affinity acquisition module is used for acquiring relevant city number and coordinate data, initializing an antibody candidate set, calculating the affinity between each antibody and an antigen, and initializing and marking each antibody;
the clone operation module is used for sequencing the calculated affinities of the antigens and the antibodies, selecting the first n antibodies with the highest affinities, carrying out clone operation on each antibody and updating the antibody marks;
the antibody marking operation module is used for marking the real-time state of the antibody after a specific step;
recalculating affinity of the mutated antibodies, selecting N antibodies with highest affinity with the antigen, putting the N antibodies into a candidate set, and performing antibody labeling updating operation on the antibodies in the candidate set;
the antibody forgetting threshold interpretation module is used for judging whether the antibody meets an antibody forgetting threshold or not, and if so, executing forgetting operation; if not, judging whether the algorithm meets the evaluation function calling times, and if so, outputting an optimal path, an optimal path graph and an optimal path distance; if not, cloning is performed for each antibody.
It is another object of the present invention to provide a program storage medium for receiving a user input, the stored computer program causing an electronic device to execute the vehicle path planning method.
It is another object of the present invention to provide a computer program product stored on a computer readable medium, comprising a computer readable program for providing a user input interface to implement the vehicle path planning method when executed on an electronic device.
By combining all the technical schemes, the invention has the advantages and positive effects that:
the invention intervenes the replacement process in the clone selection algorithm by selective forgetting, and carries out population marking and updating on the antibody in the antibody set, thereby achieving the effect of improving the convergence speed and the convergence stability of the algorithm.
The invention provides a method for solving vehicle path planning, wherein only a small amount of antibodies with high antigen affinity in an antibody candidate set generated by a clonal selection algorithm have the chance of obtaining high-frequency mutation, part of the low-affinity antibodies in the rest of the antibodies are replaced by new antibodies to participate in the next clonal selection, a large amount of antibodies with general affinity are difficult to participate in the clonal selection and exist in the antibody set for a long time, and the part of inactive antibodies are difficult to remove and update in time, so that the speed of the algorithm approaching the optimal solution is influenced.
The present invention is directed to intermediate antibodies (each abiAn urban route that a vehicle passes) and a dynamic self-adaptive forgetting method is provided. By dynamically improving the competitive potential of the new antibody and periodically forgetting the antibody with weak competitiveness, the solving precision of the clonal selection algorithm in solving the vehicle path planning problem is improved. In the aspect of algorithm convergence, the problem that the traditional clone selection algorithm is converged too early and easily falls into a local optimal solution is solved. Experiments prove that the method provided by the invention well overcomes the defects of a clone selection algorithm.
The invention is inspired by a biological forgetting mechanism, and provides an improved clone selection algorithm based on a forgetting mechanism to solve the problem of vehicle path planning. The substitution process in the clone selection algorithm is intervened by selective forgetting, and the antibodies in the antibody set are subjected to population marking and updating, so that the effects of improving the convergence speed and the convergence stability of the algorithm are achieved.
Compared with the traditional method for solving the vehicle path planning problem based on the clonal selection algorithm, the method has the following advantages that:
(1) the method introduces the ideas of the attenuation theory and the interference theory in biological forgetting into the replacement process of the clonal selection algorithm, has dynamic self-adaptability, can effectively solve the problem that the convergence of the algorithm is influenced by the intermediate antibody compared with the traditional method for solving the vehicle path planning problem based on the clonal selection algorithm, and improves the algorithm efficiency in solving the vehicle path planning problem by the clonal selection algorithm.
(2) According to the invention, the solution precision of the clonal selection algorithm in solving the vehicle path planning problem is improved by dynamically improving the competitive potential of the new antibody in the clonal selection process and periodically forgetting the antibody with weak competitiveness.
Compared with the prior art, the invention has the following technical effects or experimental effects:
compared with the existing Genetic Algorithm (GA) and Clone Selection Algorithm (CSA), the Invention (ICSA) carries out comparative experiment on the aspect of solving the vehicle path planning problem, and uses average solution error rate (MER) as a method to evaluateThe price criteria are the standard of the price,
wherein avg _ result is the average result obtained by 100 experiments, and opt is the theoretical optimal result. The closer the average result of 100 experiments is to the optimal result, the smaller the MER value is, which indicates that the more accurate the experimental result of the method is, the better the planned feasible path is.
The comparison results of the mean solution error rate (MER) values of the three methods at different transportation site scales are compared as shown in fig. 5, 6, and 7, respectively. The three experimental data sets used therein, which are classical public data sets in the field of path planning, were Bayg29 containing 29 transportation sites, Att48 containing 48 transportation sites, Eil101 containing 101 transportation sites. Experimental results prove that the method has smaller average solution error rate and can more accurately plan the optimal path for vehicle running.
Drawings
Fig. 1 is a flowchart of a vehicle path planning method according to an embodiment of the present invention.
Fig. 2 is a flowchart of a clone selection algorithm based on a forgetting mechanism according to an embodiment of the present invention.
Fig. 3 is a schematic diagram of a local crossing problem in a city according to an embodiment of the present invention.
Fig. 4 is a graph of the optimal path output provided by an embodiment of the present invention.
Fig. 5 is a comparison graph of experimental results of small-scale path planning of the vehicle path planning method provided by the embodiment of the invention and other vehicle path planning methods.
Fig. 6 is a comparison graph of experimental results of the vehicle path planning method provided by the embodiment of the invention and other vehicle path planning methods in medium-scale path planning.
Fig. 7 is a comparison graph of experimental results of the vehicle path planning method provided by the embodiment of the invention and other vehicle path planning methods in large-scale path planning.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail with reference to the following embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
In view of the problems in the prior art, the present invention provides a vehicle path planning method, which is described in detail below with reference to the accompanying drawings.
The vehicle path planning method provided by the embodiment of the invention comprises the following steps: calculating the affinity between each antibody and the antigen by initializing a candidate set of antibodies, and labeling the initial antibodies; selecting the first n antibodies with highest affinity, performing cloning operation on each antibody, and updating the selected antibody markers; carrying out mutation operation on the cloned antibody, and updating the mutated antibody mark; recalculating affinity of the mutated antibodies, selecting N antibodies with highest affinity with the antigen, and performing antibody labeling operation on the antibodies in the antibody candidate set; when the antibody meets an antibody forgetting threshold, forgetting; and terminating the calculation when the number of times of calling the evaluation function is met, and planning the vehicle path.
As shown in fig. 1-2, a vehicle path planning method provided by the embodiment of the invention includes the following steps:
s101, obtaining city related number and coordinate data, initializing an antibody candidate set, calculating affinity between each antibody and an antigen, and initializing and marking the antibodies, wherein the specific process of initializing and marking is to endow each antibody with an initial survival time value 1 and an antibody competition attribute value 1.
S102, sequencing the calculated affinities of the antigen and the antibody, selecting the first n antibodies with the highest affinities, cloning each antibody, and increasing the antibody competition attribute value of the antibodies by 1.
S103, mutation operation is carried out on the cloned antibody, and the mutated antibody is reinitialized and labeled.
S104, recalculating the affinity of the mutated antibodies, selecting N antibodies with the highest affinity with the antigen, and carrying out antibody labeling operation on the antibodies in the antibody candidate set, wherein the specific operation process is that the survival time value of each antibody is increased by 1.
S105, judging whether the antibody meets an antibody forgetting threshold, and if so, executing forgetting operation; if not, the process goes to step S106.
S106, judging whether the algorithm meets the evaluation function calling times, and if so, outputting an optimal path, an optimal path graph and an optimal path distance; if not, the process returns to step S102.
In step S101, the affinity calculation formula between the antibody and the antigen provided in the embodiment of the present invention is as follows:
wherein citysize is the total number of cities the vehicle will pass, xj、yjIs the longitude and latitude, ab, of city jiIs an antibody.
In step S103, the cloning provided in the embodiment of the present invention includes:
for antibody abiThe number of clones satisfies the following formula:
wherein a and b are both constants, and a >0, max _ clone is the maximum number of clones.
In step S103, the variations provided by the embodiment of the present invention include:
the higher the affinity, the lower the probability of antibody mutation, and mutation is performed by a mutation operator.
The mutation operator is:
among them, operator (sigma, ab)i) Is to antibody abiA certain city and neighbor cities with the interval of sigma are subjected to cross variation,
and sigma is taken according to the affinity of the antibody.
In step S105, the forgetting provided by the embodiment of the present invention includes:
acquiring iteration times gen of a current algorithm, and carrying out gen-round clonal variation operation on a new antibody; and marking the antibody of the mature antibody selected after the operation, judging whether the number of times of calling the evaluation function is met, outputting the shortest path and the total length of the path when the number of times of calling the evaluation function is met, and returning to the step S103 if the number of times of calling the evaluation function is not met.
In step S105, the forgetting formula provided in the embodiment of the present invention is:
wherein popsize is the antibody candidate population size, abnewIs a new antibody randomly generated from solution space, pre (-) is the antibody pre-treatment process, and c is the forgetting threshold.
The invention provides a system for planning a vehicle path, which comprises:
and the antibody and antigen affinity acquisition module is used for acquiring the city related number and coordinate data, initializing an antibody candidate set and calculating the affinity between each antibody and the antigen.
And the cloning operation module is used for sequencing the calculated affinities of the antigens and the antibodies, selecting the first n antibodies with the highest affinities and cloning each antibody.
The antibody marking operation module is used for carrying out mutation operation on the cloned antibody; and recalculating the affinity of the mutated antibodies, selecting n antibodies with the highest affinity with the antigen, and carrying out antibody labeling operation on the antibodies in the antibody candidate set. Calculating the degree of amnesia of the antibody
Among them, iteration (ab)i) Represents an antibody abiThe number of rounds participating in the iteration in the candidate antibody set, i.e. the survival time value, strength (ab)i) Represents an antibody abiIn whenCompetitive attribute values in the pre-iteration antibody candidate set.
The antibody forgetting threshold interpretation module is used for judging whether the antibody meets an antibody forgetting threshold or not, and if so, executing forgetting operation; if not, judging whether the algorithm meets the evaluation function calling times, and if so, outputting an optimal path, an optimal path graph and an optimal path distance; if not, cloning is performed for each antibody.
The technical solution of the present invention is further illustrated by the following specific examples.
Example 1: clone Selection Algorithm (CSA) step:
step 1, initializing relevant parameters such as the size, the iteration times, the clone number and the like of an antibody set, randomly selecting an antigen from the antigen set and generating a candidate antibody set, wherein the candidate antibody set consists of a memory set and a residual set.
And 2, calculating the affinity of each antibody in the candidate antibody set with the antigen, and selecting the first n antibodies with the highest affinity.
And 3, cloning the n antibodies, wherein the number of antibody clones is in positive correlation with the affinity of the antibodies with the antigen.
And 4, carrying out mutation treatment on the antibody set generated after cloning, wherein the higher the affinity is, the lower the probability of antibody mutation is.
And 5, calculating the affinity of the antibodies after mutation, selecting the antibodies with the highest affinity to compare with the antibodies in the current memory set, and selecting the antibodies with higher affinity to place in the memory set.
And 6, randomly selecting d new antibodies to replace the d antibodies with the worst affinity to the antigen in the rest collection.
And 7, jumping to the step 2 to perform the next iteration, and terminating the algorithm when the iteration times meet termination conditions.
When the clone selection algorithm is actually used for solving the vehicle path planning problem, the affinity between the antibody and the antibody generated randomly by the algorithm is low, and the affinity between the antibody and the antibody is excellent. When the algorithm performs the antibody renewal of step 6, the randomly generated new antibody is difficult to compete as the dominant antibody with the antibody that has undergone multiple iterations. In the next round of algorithm iteration process, the new high probability of the current round is eliminated due to too late affinity sequencing, and does not actually participate in the clone selection process of the algorithm, so that the execution efficiency of the algorithm is reduced.
In summary, the clonal selection algorithm has two problems in solving the vehicle path planning problem: firstly, the intermediate antibody cannot be eliminated in time to influence the efficiency of the algorithm, and secondly, the newly generated antibody has insufficient competitiveness when the diversity of the antibody is ensured.
Therefore, the invention is inspired by a forgetting mechanism, and designs a clone selection algorithm with antibody labeling and forgetting processes. In this model, each antibody abiIs an urban route on which vehicles run. The iteration process of the algorithm can be divided into two parts according to steps: the antibody selection mutation and labeling process and the antibody selection forgetting and updating process are adopted.
The Improved Clone Selection Algorithm (ICSA) provided by the invention comprises the following steps:
inputting: size N of candidate set of antibody, number of clones NcVariation rate r, forgetting threshold c
And (3) outputting: optimal antibodies
1 initializing a candidate set of antibodies and calculating the affinity between each antibody and an antigen;
2, selecting the first n antibodies with highest affinity, and cloning each antibody;
3, carrying out mutation operation on the cloned antibody;
4 recalculating the affinity of the mutated antibodies, selecting n antibodies with the highest affinity with the antigen, and carrying out antibody labeling operation on the antibodies in the antibody candidate set;
5 when the antibody meets the antibody forgetting threshold, executing forgetting operation;
6 when the algorithm meets the evaluation function calling times, the algorithm is stopped, otherwise, the step 2 is returned.
Example 2: the method comprises the following specific operations of solving a vehicle path planning problem by applying an improved clonal selection algorithm:
first, the inputs to the algorithm are provided: the number of cities and the coordinates of each city.
Step 1, initializing a population and setting initialization parameters;
in the present invention, an antigen means the sum of distances between all cities to be traveled by a vehicle, and an antibody means the ordered sequence of all city paths to be traveled by a vehicle. 100 path-ordered sequences containing all cities were randomly generated as the initial antibody population. The initialization parameter settings are as follows for different data sets:
table 1 initialization parameter setting table
Step 2, affinity calculation
For the vehicle path planning problem, the calculation formula of the affinity between the antibody and the antigen is as follows:
wherein citysize is the total number of cities the vehicle will pass, xj、yjIs the longitude and latitude, ab, of city jiIs an antibody.
Step 3, cloning operation
And (4) aiming at the affinity of the antigen and the antibody calculated in the last step, sequencing the affinity, and selecting the first n antibodies with high affinity in the current candidate antibody set for cloning. The greater the number of antibody clones with higher affinity, to the antibody abiThe number of clones satisfies the following formula:
wherein a and b are both constants, and a >0, max _ clone is the maximum number of clones.
Step 4, mutation operation
In the mutation operation, the higher the affinity, the smaller the probability of antibody mutation, and the invention provides a mutation operator, which is represented by the following formula:
among them, operator (sigma, ab)i) Is to antibody abiA certain city and neighbor cities with the interval of sigma are subjected to cross variation,
σ is taken as a value according to the magnitude of the affinity of the antibody, and when the affinity of the antibody is sufficiently large, σ becomes 1, and the problem of local crossover as shown in fig. 3 can be dealt with.
As shown in the left side of fig. 3, the route sequence belongs to the continuity of four locally continuous cities, and after antibody mutation, i.e. the intersection of the city q and the neighboring city m spaced by 1, the local route shown in the right side of the upper figure can be obtained, and obviously, the route after mutation is better than the route before mutation.
Step 5, forgetting operation
The significance of forgetting operation is that the diversity of antibodies is improved by a self-adaptive method, and the convergence speed of the algorithm is improved. The method judges the necessity of forgetting the antibody according to the degradation degree of the antibody.
The concrete forgetting formula is as follows:
wherein popsize is the antibody candidate population size, abnewIs a new antibody randomly generated from solution space, pre (-) is the antibody pre-treatment process, and c is the forgetting threshold. Calculating the degree of amnesia of the antibody
Among them, iteration (ab)i) Represents an antibody abiThe number of rounds participating in the iteration in the candidate antibody set, i.e. the survival time value, strength (ab)i) Represents an antibody abiA value of a competitive attribute in the current iteration antibody candidate set.
The pre-treatment of limited mutations is performed on the new antibodies generated by the forgetting manipulation, with the aim of improving the competitiveness of the new antibodies. The specific operation steps are as follows: and acquiring iteration times gen of the current algorithm, and carrying out gen-round clonal variation operation on the new antibody. And finally, carrying out antibody marking on the mature antibody selected after the operation, when the algorithm meets the number of times of calling the evaluation function, terminating the algorithm and outputting the shortest path and the total length of the path, and otherwise, returning to the step 3.
The flow chart of vehicle path planning by the clonal selection algorithm based on the forgetting mechanism is shown in fig. 1-2.
The technical effects of the present invention will be further explained below with reference to experiments.
Compared with the prior Genetic Algorithm (GA) and Clone Selection Algorithm (CSA), the Invention (ICSA) carries out comparative experiments on the aspect of solving the vehicle path planning problem, uses average solution error rate (MER) as a method evaluation standard,
wherein avg _ result is the average result obtained by 100 experiments, and opt is the theoretical optimal result. The closer the average result of 100 experiments is to the optimal result, the smaller the MER value is, which indicates that the more accurate the experimental result of the method is, the better the planned feasible path is.
The comparison results of the mean solution error rate (MER) values of the three methods at different transportation site scales are compared as shown in fig. 5, 6, and 7, respectively. The three experimental data sets used therein, which are classical public data sets in the field of path planning, were Bayg29 containing 29 transportation sites, Att48 containing 48 transportation sites, Eil101 containing 101 transportation sites. Experimental results prove that the method has smaller average solution error rate and can more accurately plan the optimal path for vehicle running. Fig. 529 MER values for the transport site scale. Fig. 648 MER value comparisons at the transport site scale. FIG. 7101 MER values comparison at shipping site scale.
It should be noted that the embodiments of the present invention can be realized by hardware, software, or a combination of software and hardware. The hardware portion may be implemented using dedicated logic; the software portions may be stored in a memory and executed by a suitable instruction execution system, such as a microprocessor or specially designed hardware. Those skilled in the art will appreciate that the apparatus and methods described above may be implemented using computer executable instructions and/or embodied in processor control code, such code being provided on a carrier medium such as a disk, CD-or DVD-ROM, programmable memory such as read only memory (firmware), or a data carrier such as an optical or electronic signal carrier, for example. The apparatus and modules thereof of the present invention may be implemented by hardware circuits such as very large scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, or programmable hardware devices such as field programmable gate arrays, programmable logic devices, or software executed by various types of processors, or a combination of the above hardware circuits and software, e.g., firmware
The above description is only for the purpose of illustrating the present invention and the appended claims are not to be construed as limiting the scope of the invention, which is intended to cover all modifications, equivalents and improvements that are within the spirit and scope of the invention as defined by the appended claims.
Claims (6)
1. A vehicle path planning method, characterized by comprising: initializing N random paths as antibodies to be placed in an antibody candidate set, calculating the distance of each vehicle path as the affinity between each antibody and an antigen, and initializing and marking each antibody; selecting the first n antibodies with highest affinity, cloning each antibody, and updating the marker; re-labeling the mutated antibody;
recalculating affinity of the mutated antibodies, selecting N antibodies with highest affinity with the antigen, and updating antibody markers of the antibodies in the antibody candidate set; when the antibody meets an antibody forgetting threshold, forgetting; terminating the calculation when the number of times of calling the evaluation function is met to obtain an optimal vehicle path;
the vehicle path planning method specifically comprises the following steps:
acquiring relevant city number and coordinate data, initializing N antibodies, putting the N antibodies into an antibody candidate set, calculating the affinity between each antibody and an antigen, and initializing a marker;
step two, ordering the calculated affinities of the antigen and the antibody, selecting the first n antibodies with the highest affinities, cloning each antibody, and updating and marking the cloned antibodies;
step three, carrying out mutation operation on the cloned antibody, and updating and marking the mutated antibody;
step four, recalculating the affinity of the mutated antibodies, selecting N antibodies with the highest affinity with the antigen, putting the N antibodies into an antibody candidate set, and adding 1 to the survival time values of the N antibodies;
judging whether the antibody meets an antibody forgetting threshold, and if so, executing forgetting operation; if not, turning to the sixth step;
the forgetting includes:
the forgetting judgment is carried out, the forgetting degree of the antibody is calculated according to the current labeling state of the antibody, the antibody is compared with a forgetting threshold value, and when the antibody meets the forgetting judgment condition, the antibody is replaced by a new antibody; acquiring iteration times gen of a current algorithm, and carrying out gen-round clonal variation operation on a new antibody; carrying out antibody initialization marking on the mature antibody selected after the operation;
the forgetting formula is:
wherein popsize is the antibody candidate population size, abnewIs a new antibody randomly generated from a solution space, pre (-) is an antibody pretreatment process, and c is a forgetting threshold; calculating the degree of amnesia of the antibody
Among them, iteration (ab)i) Represents an antibody abiThe number of rounds participating in the iteration in the candidate antibody set, i.e. the survival time value, strength (ab)i) Represents an antibody abiA competition attribute value in the current iteration antibody candidate set;
judging whether the algorithm meets the evaluation function calling times, and if so, outputting an optimal path, an optimal path graph and an optimal path distance; if not, returning to the step two.
2. The vehicle path planning method according to claim 1, wherein in the first step, the affinity between the antibody and the antigen is calculated as follows:
wherein citysize is the total number of cities the vehicle will pass, xj、yjIs the longitude and latitude, ab, of city jiIs an antibody;
the specific process of initializing the labeling in step one is to assign an initial survival time value of 1 and an antibody competition attribute value of 1 to each antibody.
3. The vehicle path planning method according to claim 1, wherein in step two, the cloning comprises:
for antibody abiThe number of clones satisfies the following formula:
wherein a and b are constants, a is greater than 0, and max _ clone is the maximum clone number;
and the specific process of updating the marker in the second step is to add 1 to the competitive attribute value of the cloned antibody.
4. The vehicle routing method of claim 1, wherein in step three, the mutating comprises:
the higher the affinity, the smaller the probability of antibody mutation, and mutation is carried out by using a mutation operator;
the mutation operator is:
among them, operator (sigma, ab)i) Is to antibody abiA certain city and neighbor cities with the interval of sigma are subjected to cross variation,
carrying out value taking on sigma according to the size of the affinity of the antibody;
the specific process of updating the marker in step three is to assign an initial survival time value of 1 and an antibody competition attribute value of 1 to the antibody generating the variation.
5. A vehicle path planning system using the vehicle path planning method according to any one of claims 1 to 4, the vehicle path planning system comprising:
the antibody and antigen affinity acquisition module is used for acquiring relevant city number and coordinate data, initializing an antibody candidate set, calculating the affinity between each antibody and an antigen, and initializing and marking each antibody;
the clone operation module is used for sequencing the calculated affinities of the antigens and the antibodies, selecting the first n antibodies with the highest affinities, carrying out clone operation on each antibody and updating the antibody marks;
the antibody marking operation module is used for marking the real-time state of the antibody;
recalculating affinity of the mutated antibodies, selecting N antibodies with highest affinity with the antigen, putting the N antibodies into a candidate set, and performing antibody labeling updating operation on the antibodies in the candidate set;
the antibody forgetting threshold interpretation module is used for judging whether the antibody meets an antibody forgetting threshold or not, and if so, executing forgetting operation; if not, judging whether the algorithm meets the evaluation function calling times, and if so, outputting an optimal path, an optimal path graph and an optimal path distance; if not, cloning is performed for each antibody.
6. A program storage medium receiving a user input, the stored computer program causing an electronic device to perform the vehicle path planning method of any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010176733.6A CN111337042B (en) | 2020-03-13 | 2020-03-13 | Vehicle path planning method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010176733.6A CN111337042B (en) | 2020-03-13 | 2020-03-13 | Vehicle path planning method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111337042A CN111337042A (en) | 2020-06-26 |
| CN111337042B true CN111337042B (en) | 2021-11-02 |
Family
ID=71180024
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010176733.6A Active CN111337042B (en) | 2020-03-13 | 2020-03-13 | Vehicle path planning method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111337042B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114781966B (en) * | 2022-04-08 | 2024-04-12 | 重庆大学 | Logistics distribution path planning method, device, equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103412490A (en) * | 2013-08-14 | 2013-11-27 | 山东大学 | Polyclone artificial immunity network algorithm for multirobot dynamic path planning |
| CN104407616A (en) * | 2014-12-03 | 2015-03-11 | 沈阳工业大学 | Dynamic path planning method for mobile robot based on immune network algorithm |
| CN107578132A (en) * | 2017-09-11 | 2018-01-12 | 天津工业大学 | A kind of garbage truck method for obtaining path based on immune optimization |
| CN109597924A (en) * | 2018-09-14 | 2019-04-09 | 湖北大学 | A kind of microblogging social circle method for digging and system based on artificial immune network |
| CN110362107A (en) * | 2019-06-17 | 2019-10-22 | 杭州电子科技大学 | The method for solving of multiple no-manned plane Task Allocation Problem based on immune optimization algorithm |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| AU777041B2 (en) * | 1999-05-17 | 2004-09-30 | Crucell Holland B.V. | Adenovirus derived gene delivery vehicles comprising at least one element of adenovirus type 35 |
| US20160321757A1 (en) * | 2015-04-29 | 2016-11-03 | Pacific Resources Benefits Advisors, Llc | Methods and systems for generating, and responding to, requests for proposals and requests for information for insurance products |
| CN105550779B (en) * | 2015-12-25 | 2019-06-25 | 武汉理工大学 | A kind of large-scale cable machine GPS guided path optimization method based on immune algorithm |
| CN107567005B (en) * | 2017-06-12 | 2021-01-01 | 华东师范大学 | Internet of vehicles abnormal behavior detection method and system based on artificial immune system |
| CN110516884A (en) * | 2019-08-30 | 2019-11-29 | 贵州大学 | A kind of short-term load forecasting method based on big data platform |
-
2020
- 2020-03-13 CN CN202010176733.6A patent/CN111337042B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103412490A (en) * | 2013-08-14 | 2013-11-27 | 山东大学 | Polyclone artificial immunity network algorithm for multirobot dynamic path planning |
| CN104407616A (en) * | 2014-12-03 | 2015-03-11 | 沈阳工业大学 | Dynamic path planning method for mobile robot based on immune network algorithm |
| CN107578132A (en) * | 2017-09-11 | 2018-01-12 | 天津工业大学 | A kind of garbage truck method for obtaining path based on immune optimization |
| CN109597924A (en) * | 2018-09-14 | 2019-04-09 | 湖北大学 | A kind of microblogging social circle method for digging and system based on artificial immune network |
| CN110362107A (en) * | 2019-06-17 | 2019-10-22 | 杭州电子科技大学 | The method for solving of multiple no-manned plane Task Allocation Problem based on immune optimization algorithm |
Non-Patent Citations (4)
| Title |
|---|
| An Improved Clonal Selection Algorithm and its Application in Function Optimization Problems;Yidan Luo 等;《Second International Symposium on Intelligent Information Technology Application》;20081231;第118-121页 * |
| Experimenting Dynamic Clonal Selection (DCS) for Parallel Multiple Interest Topics of User Profile Adaptation in Content Based Filtering;Nurulhuda Firdaus Mohd Azmi 等;《Procedia Computer Science》;20191231;第161卷;第433-440页 * |
| 基于免疫遗忘的车间调度算法;郭江平;《计算机仿真》;20100430;第27卷(第4期);第317-332页 * |
| 基于改进克隆选择算法的仓储布局优化设计;曹霞 等;《控制工程》;20200229;第27卷(第2期);第329-334页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111337042A (en) | 2020-06-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107168054B (en) | Multi-robot task allocation and path planning method | |
| Zhang et al. | Bottleneck machine identification method based on constraint transformation for job shop scheduling with genetic algorithm | |
| CN109271320A (en) | A kind of upper multiple target priorities of test cases sort method | |
| CN112013829A (en) | Multi-UAV/UGV (unmanned aerial vehicle/user generated Unit) cooperative long-term operation path planning method based on multi-objective optimization | |
| CN111337042B (en) | Vehicle path planning method and system | |
| Wang et al. | Modeling and planning multimodal transport paths for risk and energy efficiency using and/or graphs and discrete ant colony optimization | |
| CN110334853A (en) | A kind of imitative nature body optimization method of logistics distribution center Warehouse Location | |
| Gould et al. | A material flow modelling tool for resource efficient production planning in multi-product manufacturing systems | |
| Yuan et al. | Dual-resource integrated scheduling method of AGV and machine in intelligent manufacturing job shop | |
| CN113341889A (en) | Distributed blocking flow workshop scheduling method and system with assembly stage and energy consumption | |
| Kalakanti et al. | A hybrid cooperative method with Lévy flights for electric vehicle charge scheduling | |
| Lin et al. | Niching Pareto ant colony optimization algorithm for bi-objective pathfinding problem | |
| Wu et al. | Real environment-aware multisource data-associated cold chain logistics scheduling: A multiple population-based multiobjective ant colony system approach | |
| Cao et al. | Surrogate-assisted symbiotic organisms search algorithm for parallel batch processor scheduling | |
| Xiao et al. | A CEGA-based optimization approach for integrated designing of a unidirectional guide-path network and scheduling of AGVs | |
| CN111340303A (en) | Route planning method for travelers based on novel mixed frog-leaping algorithm | |
| Liu et al. | An estimation of distribution algorithm with multiple intensification strategies for two-stage hybrid flow-shop scheduling problem with sequence-dependent setup time | |
| Cheng et al. | Mathematical model and adaptive simulated annealing algorithm for mixed-model assembly job-shop scheduling with lot streaming | |
| Wu et al. | Neural airport ground handling | |
| Tan et al. | Multi-objective teaching–learning-based optimization algorithm for carbon-efficient integrated scheduling of distributed production and distribution considering shared transportation resource | |
| Liu et al. | A Cooperative Evolutionary Computation Algorithm for Dynamic Multiobjective Multi-AUV Path Planning | |
| Duan et al. | EDA based probabilistic Memetic Algorithm for distributed blocking permutation flowshop scheduling with sequence dependent setup time | |
| Chen | An improved genetic algorithm for solving the Traveling Salesman Problem | |
| Yusof et al. | Constraint-chromosome genetic algorithm for flexible manufacturing system machine-loading problem | |
| CN110705650A (en) | Metal plate layout method based on deep learning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |